Stuck on picking a World Cup winner? Data analytics has the answer

Analytics

by Michael Moore| 09 May 2014

Brazil most likely to triumph, according to leading analytics firm.

Anyone hoping for a British Eurovision win this weekend or an England World Cup triumph later this summer may end up feeling slightly disappointed, as a leading data analytics firm have revealed neither will be the case.

PredictWise.com, headed up by economist David Rothschild, specialises in forecasting the most likely winner of an event, covering the outcomes of events ranging from this summer's football World Cup and Champions League, the Eurovision Song Contest this weekend, as well as upcoming political elections.

Rothschild is interested in how markets develop over time, particularly as more information becomes available from a variety of sources regarding a certain event.

"Things don't surprise me," he says, adding that his definition of accuracy is "probably different to most people".

So far, he has Brazil as the most likely World Cup winners (with a 24% likelihood of winning, ahead of Argentina and Germany), Real Madrid to win the Champions League (with a 47.6% likelihood of success), and Sweden to win this weekend's Eurovision (with a 24.7% success rate).

The likelihoods of all these events are calculated using a selection of sources providing continuously updating data, varying by the various outcomes he is looking to predict. For example, in looking to predict the winner of a football match, Rothschild will look at previous win/loss records, as well as statistics from current games, but will also examine the markets surrounding the sport as a whole, which could mean looking at betting patterns or even chatter from social media.

But as well as upcoming sporting events, Predictwise.com also contains predictions concerning more long-term events, including the 2016 US Presidential election, which often use what Rothschild calls 'fundamental data', which can include past trends and election results, along with economic indicators.

"Time is a great variant," says Rothschild, stating that he often uses several different models to calculate one outcome, especially in politics, where the markets often change greatly as a campaign progresses and the variety of possible outcomes starts to shrink.

It's a great time to get into forecasting, Rothschild believes, as "so much is shifting". His data has changed, he says, "from a world of static, non very responsive indicators, that are generally sometimes very qualitative, to a world that's very quickly updating high quantitative indicators."

But that's not to say that his predictions are always true, however, as he admits that far from predicting the future, often he is just forecasting what the most likely turn of events is.

"Sometimes, it doesn't matter how much information I have," he says. "There is always going to be probabilities - and sometimes they're not that high, but you want to make sure that, if I say something happening at 25%, I don't want it happening ten times out of ten, I want it happening 2.5 times out of ten."

Comments
Post a comment

Comments may be moderated for spam, obscenities or defamation.

Join our network

753 people like this.
0 people follow this.

Analytics Intelligence

Buy the latest industry research online today!
See more
Privcy Policy

We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.